The whole web, in one API call
The web-data layer for AI agents.
Real Synoppy output — refreshed daily, one key for every product.
Everything the web can give an agent
One key, one base URL — every capability below is a single POST to the endpoint shown on its card. Grouped fields (like a full brand profile) come back in one call.
Read
POST /api/scrapeAny URL → clean markdown, HTML, or text.
- Markdown
- HTML
- Plain text
- JS rendering
Crawl
POST /api/crawlRead every page on a domain.
- Whole-site crawl
- Markdown per page
- Change monitoring
Map
POST /api/mapEvery URL on a domain, fast.
- URL discovery
- sitemap.xml parsing
- Same-origin link crawl
Extract
POST /api/extractPages → structured JSON, with AI.
- Structured JSON
- Your schema or prompt
Images
POST /api/imagesEvery image on a page.
- All images + alt text
- Lazy-load & srcset
Screenshot
POST /api/screenshotPixel-perfect page captures.
- Viewport or full page
- PNG output
Enrich
POST /api/brandAny domain → full brand profile, one call.
- Company logos
- Brand colors
- Fonts
- Socials
- Description
- Address
- Full profile
Classify
POST /api/classifyPages → business signals.
- NAICS code
- SIC code
- Custom labels
Act
Coming soonAgents that complete real tasks.
- Run workflow
- Fill forms
- Authenticated sessions
Real output, zero setup
One call in, clean structured data out. These are real API responses, refreshed daily — not screenshots.
## Product
## Stripe Projects adds new agent inte…
June 11, 2026
Rami Banna Product Lead, Stripe Projects
{
"title": "Hacker News",
"summary": "This page displays a curated list of trendi…,
"topics": ["Artificial Intelligence","Software Enginee…,
"entities": ["Hacker News","Meta","Claude","Moderna","FD…
}Live in three steps
From signup to clean, structured data in minutes — no infrastructure to run.
Get your API key
Sign up free in seconds — no credit card. You start with 1,000 credits, enough to test every format on real pages.
Call the API
One POST request. Point it at any URL and choose the format — markdown, HTML, JSON, or a screenshot.
Get clean data
Receive LLM-ready markdown or structured JSON. No boilerplate, no markup noise — just the content your model needs.
curl -X POST https://synoppy.com/api/scrape \
-H "Authorization: Bearer $SYNOPPY_KEY" \
-H "Content-Type: application/json" \
-d '{ "url": "example.com", "formats": ["markdown"] }'The easiest web data to drop in
One key, typed SDKs, and an MCP server. No proxies, no headless browsers, no scraping infra to babysit.
- 1
Get your key
Sign up free and grab an API key from the dashboard. 1,000 credits included.
- 2
Install the SDK
One line in TypeScript, Python, Go, or Ruby. Or skip it and call the REST API.
- 3
Make your first call
One function, clean markdown back. Or wire the MCP server into Claude or Cursor.
npm install @synoppy/sdkconst doc = await syn.read("stripe.com/blog");We handle the hard stuff
Retries, clean extraction, token-efficient output, and live JS rendering — with anti-bot rolling out — so your agents just get data.
JS rendering
Full headless rendering returns the page a real user sees — not an empty shell. Live on the same endpoint: set render:true or render:"auto".
Anti-bot and proxies — on the roadmap
Rotating proxies and stealthy sessions to get through the defenses that block naive scrapers — coming soon.
Smart waits and retries
We wait for content to settle and retry transient failures automatically, so flaky pages still come back clean.
Clean extraction
Nav bars, ads, and boilerplate are stripped out — leaving just the content your agents actually need.
Rate limits handled
Concurrency, backoff, and queuing are managed for you, so you can scale without babysitting throttles.
Token-efficient output
Tight, structured markdown means fewer tokens per page and lower cost in every downstream model call.
From a single read to the whole web
Teams use Synoppy as the data layer underneath agents, pipelines, and products — every use case below runs on the live API today.
RAG & knowledge bases
Feed agents fresh, clean markdown instead of brittle HTML. Build retrieval over any site and keep it current on a schedule.
Competitive & price monitoring
Pull pricing, listings, and copy across thousands of pages, re-run on a schedule, and diff the markdown for changes.
Lead & company enrichment
Resolve a domain into brand assets, fonts, socials, and an industry classification — typed, in one or two calls.
Structured data extraction
Turn any page into typed JSON your app or agent can use — define the shape in a prompt and let the model fill it.
Site mapping & audits
Discover every URL on a domain and pull its images to audit structure, coverage, and assets at scale.
Content & docs pipelines
Convert any source into token-efficient markdown for summarization, translation, and ingestion downstream.
Pay only for what you run
Credits scale with the operation and pool across every product. No infrastructure to run.
Free
Prototype and kick the tires.
- 1,000 credits / mo
- 2 requests / sec
- All live products
- Community support
Starter
For indie devs shipping a real product.
- 5,000 credits / mo
- 10 requests / sec
- All live products
- Email support
Pro
Scaling agents and data pipelines.
- 100,000 credits / mo
- 50 requests / sec
- All live products
- Webhooks
- Priority support
Scale
High-volume production workloads.
- 1,000,000 credits / mo
- 150 requests / sec
- Production-grade availability
- Slack support channel
Enterprise
Custom credit volume, dedicated capacity, invoicing, SLAs, and a dedicated account manager.
Questions, answered
Read (any URL into clean markdown, HTML, or text with full metadata), Crawl, Map, Images, Enrich, AI Extract, Classify, JS rendering, and screenshots are all live today on one API key and one metered credit balance. Act (agent automation) is coming soon.
Give your agents the whole web
Read, crawl, map, extract, enrich, classify, and images are live today — all on one key. Agent actions are on the way. Build the next thing on Synoppy.